Tight feasibility thresholds are derived for the minimal sub-optimality gap in convex L-smooth distributed optimization under bounded adversarial gradient perturbations, together with algorithms attaining them at matching query complexity.
Learning rates for stochastic gradient descent with nonconvex objectives.IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 43(12):4505–4511, 2021
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Distributed Learning with Adversarial Gradient Perturbations
Tight feasibility thresholds are derived for the minimal sub-optimality gap in convex L-smooth distributed optimization under bounded adversarial gradient perturbations, together with algorithms attaining them at matching query complexity.